Endoscopy is an invasive procedure in which a flexible optical fiber (the endoscope) is used for visually inspecting the inner lining of organs such as the airways (bronchoscopy) or the colon (colonoscopy). Virtual endoscopy (VE) uses computed tomography (CT) or magnetic resonance (MR) data to reconstruct and display airway (VB) or colon (VC) surfaces on computers, thereby eliminating the need for the insertion of an endoscope. To date, CT or MR-based virtual endoscopy only examines the surface of the structures. Positron emission tomography (PET) with fluoro-deoxy-glucose (FDG; an analog of glucose) or other tracers can characterize cancerous tissues based on the differentiated metabolism (glycolytic activity for FDG) they exhibit. By combining PET with CT data, we can create a novel modality: PET-augmented virtual endoscopy (PaVE), that offers the anatomical definition provided by CT, but also enables further evaluation for possible malignant involvement based on the glycolytic activity (with FDG) exhibited through PET. Through the clinical integration of new optimized protocols and software tools in a multidisciplinary environment, we aim to improve our ability to effectively combine PET and CT data sets in a VE paradigm with the prospect of increasing diagnostic accuracy and of guiding minimally invasive (Ml) tissue staging procedures such as bronchoscopic transbronchial biopsies and tumor resections for cancers of the chest and lungs. Recently introduced combination PET/CT units that yield both PET and CT data sets in a single study will be instrumental to our proposal. Our institution has been among the very first ones to pioneer the application of these hybrid machines to cancer diagnosis and management, and to further expand their use and accuracy through respiratory gating. The current proposal is therefore a natural complement and extension of this endeavor that conforms to the general purpose of PA-04-045 and especially to topics 3 (image displays and analyses) and 4 (image-guided cancer intervention). [unreadable] [unreadable] [unreadable]